This repository contains scripts that demonstrates different ways to interact with the rest.uniprot.org API.
Python 3
git clone https://github.com/ebi-uniprot/uniprot-scripts.git
cd uniprot-scripts
pip3 install requests
This script queries UniProtKB for features with type, description and length conditions all satisfied at the same time. It can be called from the command line as seen in the following example:
./python/get_accessions_with_feature_type_description_length.py \
--ft_type "motif" \
--ft_description "9aaTAD" \
--ft_min_length 10 \
--ft_max_length "*" \
--and_query "(organism_id:9606)" \
--out results.list
The saved results can then be used as part of a Retrieve/ID mapping job to retreive additional data (eg FASTA, TSV, JSON)
You can use search.py as a template for writing your own scripts or just modify the main() function with your own parameters. This file contains two different ways to perform a search:
search_pagination(query, size=100, namespace="uniprotkb")
This way of searching uses the search endpoint which retrieves the results in batches. This can be useful when you expect the query to be large and you want to apply some function on each batch as it arrives.
search_stream(query, namespace="uniprotkb")
This way of searching used the stream endpoint which fetches everything at once. This can be useful if you just want all of the result at once. If there is a disruption in your connection this will mean you will need to restart the stream search so search can be more robust in this respect.